Pull the text out of a scanned PDF or image-based document. The OCR engine runs entirely on your device.
A scanned PDF looks like text but is really a picture of text — you can't search or copy from it. OCR (optical character recognition) reads the image and produces real, selectable text. CorePDFTools runs OCR in your browser.
Scanned documents are often personal, so uploading them to an OCR server is a privacy concern. CorePDFTools runs the recognition on your device. OCR works best on clear, straight, reasonably high-resolution scans; faint or skewed scans produce more errors.
Turn a picture-of-text scan into real text you can search and copy.
Pull the text out of a scanned image for reuse elsewhere.
Accuracy depends on scan quality. Clean, high-contrast, straight scans give the best results.
Most OCR is built for printed text; handwriting recognition is far less reliable.
No. OCR runs entirely in your browser.
Yes, in your mobile browser, though large documents may be slower.
Many, including non-Latin scripts — pick the one matching your document for best results.